Hacker News: phreeza

New comment by phreeza in "Opus 5.5 agents discover two room-temperature magnetic semiconductor candidates"

phreeza — Tue, 06 Oct 2026 14:18:55 +0000

The glory of having bested a stranger oh the internet in a battle of whits is enough for me!

New comment by phreeza in "Opus 5.5 agents discover two room-temperature magnetic semiconductor candidates"

phreeza — Tue, 06 Oct 2026 14:10:11 +0000

I'll take that bet? Room temperature semiconductors are very common.

New comment by phreeza in "How Singapore's government-run dating service works"

phreeza — Fri, 02 Oct 2026 06:32:13 +0000

Nice idea in principle but (in the case of private) too long a time horizon for typical corporations to care about, sadly.

New comment by phreeza in "The Millennium Problems for Biology"

phreeza — Sun, 20 Sep 2026 16:07:30 +0000

I agree with you there are other much grander challenges, but apart from that I don't think the list is horrible, because as I said it's incentive-aligned with the current moment we are in. Your suggestion would definitely be a good addition to the list for sure.

I think some of the items on the list are also way less likely to be worked on (like cryogenics) than others, because they really require some heavy infrastructure to iterate. My money for which of these gets cracked with help of an AI would be rubisco, that can be quite effectively worked on in a closed loop lab fashion with a standard lab or CRO.

New comment by phreeza in "The Millennium Problems for Biology"

phreeza — Sun, 20 Sep 2026 15:35:19 +0000

Those don't have the property of being easily verified, which the problems proposed here do. I think that's a smart idea because it's a way to capture quick PR seeking AI-lab dollars for quite useful outcomes.

New comment by phreeza in "Why some US restaurants are banning tips"

phreeza — Tue, 25 Aug 2026 10:16:39 +0000

Next also include sales tax in label prices.

New comment by phreeza in "New Worlds: We are living in the future of J.G. Ballard or William Gibson"

phreeza — Fri, 21 Aug 2026 14:39:27 +0000

I agree, it is easy to miss just because we get used to things quickly. I remember this thought hitting me vividly when I saw this picture during COVID https://cdn.theatlantic.com/thumbor/MUBajBy-350E5JAjtntEhAYz...

New comment by phreeza in "There Will Come Soft Rains (1950) [pdf]"

phreeza — Tue, 04 Aug 2026 20:47:14 +0000

Funny there the date is April 1985, a lot earlier and almost my birthday!

New comment by phreeza in "Are AI labs pelicanmaxxing?"

phreeza — Thu, 23 Jul 2026 09:06:33 +0000

I think a lot of LLM progress in the last 2 years or so is really from densifying the training data to niches like this and not so much from general problem solving capability. I guess the hope is that the deluge of super specific training data will pay off through better overall competence. cf https://www.dwarkesh.com/p/the-sample-efficiency-black-hole

New comment by phreeza in "How we measured AI writing across arXiv, and where the measurement breaks"

phreeza — Mon, 20 Jul 2026 17:10:09 +0000

This is a pretty stunning result. The time series looks really convincing. Is the way the detector itself is trained orthogonal to this or could there be some "leakage" in that the pre-chatgpt text is in the (positive) training data?

Locals and Tourists

phreeza — Thu, 16 Jul 2026 05:10:15 +0000

Article URL: https://www.flickr.com/photos/walkingsf/albums/72157624209158632/

Comments URL: https://news.ycombinator.com/item?id=48930601

Points: 5

# Comments: 0

New comment by phreeza in "Interrail: 6,379Km and 13 Countries over 7 weeks"

phreeza — Mon, 13 Jul 2026 15:04:04 +0000

For Switzerland, that's only true for the special tourist train panoramic carriages on the route, there are normal trains running on the same routes, too, that don't require reservations.

New comment by phreeza in "Europe's new climate in seven charts"

phreeza — Sun, 05 Jul 2026 12:13:38 +0000

Or solar geoengineering? Which is scary but becoming more likely it seems.

New comment by phreeza in "Senior SWE-Bench: open-source benchmark that assesses agents as senior engineers"

phreeza — Thu, 02 Jul 2026 06:29:20 +0000

I think this is a complete misunderstanding of what people mean by taste in software engineering. Taste is more like the System 1 response one builds to code over time, which (ideally) captures the quality of the software beyond surface level, so things like maintainability, composability, readability, likelihood of hidden bugs. This is completely different from the question if the code fulfills the immediate task at hand, but also not the same as pure aesthetics.

New comment by phreeza in "Web Browsers on PDAS"

phreeza — Sat, 20 Jun 2026 20:13:23 +0000

Was just going to ask if anyone remembered the name of this one. It was amazing! 15 year old me felt very important/futuristic syncing the news before school in the morning and reading during school break!

Edit: after looking it up with plucker as a starting point I think the one I actually used was AvantGo, same idea though I think.

New comment by phreeza in "Police officer investigated for using AI to 'create evidence' in multiple cases"

phreeza — Sun, 14 Jun 2026 08:26:47 +0000

Maybe its more like the hash was a well known secure hash but someone managed to extract the salt/private key/signing certificate from the camera?

New comment by phreeza in "Ask HN: What was your "oh shit" moment with GenAI?"

phreeza — Sat, 06 Jun 2026 06:53:36 +0000

I ran Claude Code on my ca 2015 ThinkPad which was having wifi issues and asked it to fix them. It diagnosed the problem and applied some obscure kernel flag which fixed the issue.

New comment by phreeza in "Uber's $1,500/month AI limit is a useful signal for AI tool pricing"

phreeza — Thu, 04 Jun 2026 06:26:54 +0000

There are plenty of expenses in this order of magnitude that are not tied to direct increases in productivity. I think it may become a serious hiring impediment for companies to be really skimpy on these budgets for example.

New comment by phreeza in "Meta workers can opt out of being tracked at work up to 30 min"

phreeza — Wed, 03 Jun 2026 15:16:57 +0000

Is it really that different with the current iteration of AI compared to what was possible 10 years ago? There may be some new awareness at the executive level of what is possible, but I feel like a "slacker detector" or whatever would have been possible with xgboost or lstms.

New comment by phreeza in "AI Agent Guidelines for CS336 at Stanford"

phreeza — Mon, 01 Jun 2026 19:48:39 +0000

Pretty sure op is talking about university in Spain for other subjects, not Spanish class at university.